Intel and some of the top high-performance computing facilities in the world are testing quantum computers while they sift through various systems and methods to find the one that works best for their infrastructure.
New Services are Now Possible
The mass production of quantum chips is now at the center of a conflict over quantum superiority, which can be difficult given that qubits can be notoriously unstable and challenging to scale to the millions of qubits needed for universal quantum computers.
According to a small test, Intel claims to have fixed some problems. A significant step toward the company’s long-term objective of creating a universal quantum computer was reached when the chipmaker could produce stable quantum dots in its current factories. The company isolated and examined the uniformity of quantum dots on its wafer using a testing tool known as a “cryoprober.”
Intel has asserted an early advantage in the production of quantum systems during this process. According to analysts, Intel’s shift to manufacturing products with advanced packaging opens the door to developing more sophisticated computing systems, such as quantum devices.
AI, quantum, and Neuromorphic Computing Innovations 2022
By the end of the year, the company hopes to have early quantum hardware delivered to Argonne National Laboratory, which is testing various types of quantum hardware as part of the Q-Next program run by the U.S. Department of Energy. In its ongoing competition with China for supremacy in computing, the U.S. government is prioritizing quantum hardware. China also is advancing its own quantum computing goals.
Intel is pursuing a universal quantum computer, which experts agree will need more than a million qubits, like IBM and Google. Superconducting qubits, which can be challenging to produce using current processes, are a target for Google and IBM. The company’s quantum computing system is based on quantum dots and uses chips that can grow in Intel’s current factories.
According to David Kanter, principal analyst at Real World Technologies, for quantum to be relevant, it must be produced in large quantities. Intel has a keen understanding of that in a way that many other organizations don’t.
It will take many, many thousands, possibly even millions of raw qubits to produce fault-tolerant, stable logical qubits, according to Kevin Krewell, the principal analyst at Tirias Research. Scaling is crucial to completing this, according to Krewell.
According to Dylan Patel, the founder of SemiAnalysis, a semiconductor research and consulting company, to achieve maximum overlay performance in quantum computing, the qubits should ideally be as identical as possible. 300mm wafers and more sophisticated lithography tools have the best overlay capabilities.
IBM and D-Wave sparred over which qubit was better in the early days of quantum computing. The Jülich Supercomputing Centre in Germany uses D-quantum Wave’s annealer system for optimization, and IBM claimed that it is not a true quantum computer. Rather than being a full-fledged quantum computer, D-computer Wave is now widely recognized as a small-scale quantum computer that is effective at optimization.
According to Intel’s Clarke, universal quantum computers might make optimizers less valuable because they could be programmed for chemistry, cryptography, optimization, and other uses.
Although D-Wave and Rigetti, acquired by SPAC firms as high-risk, high-reward acquisitions, are struggling due to business realities, the qubit superiority debate is still active among the dozens of quantum computing companies. D-Wave, which counts financial, automotive, and defense companies among its clients, is already making a sizable profit from its annealer and is attempting to construct a universal quantum computer.
Different kinds of quantum computers can be used to address various problems, as recognized by Argonne and other high-performance computing facilities. These quantum computers will be attached as accelerators to powerful computers with CPUs and GPUs. Quantum hardware is a part of the service portfolios of the top three cloud providers, Amazon, Microsoft, and Google.
The ability to manufacture quantum computers in current factories and use widely accessible materials is crucial for their commercialization. Rigetti has bragged about its in-house capacity to create its quantum computer as a viable method of delivering such systems. Intel is refocusing on silicon and claims its manufacturing facilities offer a more practical route to quantum computing.
Uncertainty exists regarding whether Intel will make the quantum toolkit and advancements accessible through the Intel Developer Cloud, which was announced at the Intel Innovation show last month. The cloud service offers remote access to upcoming chips like the Gaudi 2 A.I. chip and Sapphire Rapids server chips.